INDEX

Explanations

user interfaces

np_max-act · gemini-2.0-flash

instructions related to navigating and managing digital content or software.

oai_token-act-pair · gpt-4o-mini Triggered by @xinyanhu8

The neuron detects descriptions of UI elements being highlighted or outlined in green (e.g. surrounded by a green square or turning green).

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 hotter

-0.07

!(↵

-0.06

{EIF

-0.06

Inst

-0.06

Expand

-0.06

ублі

-0.06

="""

-0.06

systems

-0.06

těž

-0.06

aims

-0.06

POSITIVE LOGITS

UTDOWN

0.06

rk

0.06

	form

0.06

,value

0.06

 это

0.06

/car

0.06

##

0.06

 leopard

0.06

्पत

0.06

Activations Density 0.020%

user interfaces

instructions related to navigating and managing digital content or software.

The neuron detects descriptions of UI elements being highlighted or outlined in green (e.g. surrounded by a green square or turning green).

No Comments

No Known Activations